DPAGait: Dynamic Pose-Knowledge-Aware Network for Robust Gait Recognition in the Wild
作者: Zenan Lin, Junduan Huang, Kun Liu, M. Saad Shakeel, Wenxiong Kang
分类: cs.CV
发布日期: 2026-11-03
💡 一句话要点
提出动态姿态知识感知网络以解决野外步态识别问题
🎯 匹配领域: 支柱六:视频提取与匹配 (Video Extraction)
关键词: 步态识别 动态姿态解析 深度学习 计算机视觉 生物特征识别 3D模型 鲁棒性 特征提取
📋 核心要点
- 现有步态识别方法依赖低信息熵的二值轮廓图像,未能有效应对野外环境中的服装、视角和动态条件变化。
- 提出动态姿态解析增强的轮廓步态表示方法,结合2D和3D特征提取,提升步态识别的准确性和鲁棒性。
- 在多个真实世界数据集上进行广泛实验,验证了所提方法的有效性,取得了竞争力的性能表现。
📝 摘要(中文)
步态生物识别旨在通过人类行走时捕获的外观或姿态特征进行身份识别。现有方法主要依赖于二值轮廓图像,信息熵低,未能充分利用步态序列中的动态信息,导致在野外环境中识别准确性和鲁棒性显著下降。为此,本文提出了一种专门设计的姿态知识感知方法,结合动态姿态解析与传统轮廓,显著增强信息熵并捕获细粒度行为线索。通过构建双分支动态姿态知识感知步态网络,整合2D动态姿态特征与3D动态姿态信息,进行准确的步态识别。实验结果表明,该方法在多个挑战性数据集上表现出色,具有良好的泛化能力。
🔬 方法详解
问题定义:本文旨在解决现有步态识别方法在野外环境中面临的挑战,包括服装变化、视角差异和动态环境条件对识别准确性的影响。现有方法主要依赖于低信息熵的二值轮廓图像,无法充分利用步态序列中的动态信息。
核心思路:提出了一种姿态知识感知的方法,通过动态姿态解析增强传统轮廓表示,结合2D和3D特征提取,旨在提高步态识别的准确性和鲁棒性。
技术框架:整体框架由两个互补的分支组成:1) 2D动态姿态感知分支,从提出的动态姿态解析增强轮廓表示中提取特征;2) 3D皮肤多人体线性分支,利用3D SMPL模型捕获丰富的动态姿态信息。最终将提取的特征进行整合,以实现准确的步态识别。
关键创新:最重要的技术创新点在于提出了动态姿态解析增强的轮廓表示方法,显著提高了信息熵,并有效捕获了细粒度的行为线索。这一方法与传统依赖二值轮廓的步态识别方法有本质区别。
关键设计:在网络结构设计上,采用双分支架构,分别处理2D和3D特征,确保信息的全面性和准确性。损失函数设计上,结合了多任务学习的策略,以优化整体识别性能。
🖼️ 关键图片
📊 实验亮点
在SUSTech1K、CCPG和Gait3D等多个真实世界数据集上进行的实验表明,所提方法在步态识别任务中表现出色,取得了竞争力的性能,尤其是在动态环境条件下,识别准确率显著提升,验证了方法的有效性和泛化能力。
🎯 应用场景
该研究在安全监控、身份验证和人机交互等领域具有广泛的应用潜力。通过提高步态识别的准确性和鲁棒性,可以有效应对现实生活中的多种挑战,推动智能监控系统和个性化服务的发展。未来,该技术有望在智能城市和公共安全等领域发挥更大作用。
📄 摘要(原文)
Gait biometrics aims at identity recognition by appearance or pose features captured during unconstrained human walking. Existing methods predominantly rely on binary silhouette images, inherently characterized by low information entropy, and fail to fully utilize the dynamic information in gait sequences, limiting their ability to handle cross-time variations effectively. Consequently, these methods encounter significant challenges in the wild. They are susceptible to variations in clothing, carried objects, and viewpoints, as well as dynamic environmental conditions, all of which lead to substantial degradation in recognition accuracy and robustness. To overcome these challenges, we propose a pose-knowledge-aware approach specifically designed for robust gait recognition in the wild. First, we propose a novel Dynamic Pose-Parsing enhanced Silhouette gait representation paradigm that substantially enhances the information entropy and effectively captures fine-grained behavioral cues by integrating pose-parsing cues with traditional silhouettes. Subsequently, we develop a dual-branch Dynamic Pose-Knowledge-Aware Gait network. The framework consists of two complementary branches: (1) a 2D dynamic pose-aware branch that extracts pose-enhanced features from our proposed DPPS representations, and (2) a 3D Skinned Multi-Person Linear branch that leverages the 3D SMPL model to capture rich and comprehensive 3D dynamic pose information. Finally, the features are integrated to facilitate accurate and robust gait recognition. Extensive experiments on four challenging in-the-wild gait datasets demonstrate the effectiveness and generalizability of the proposed DPPS representation. The complete framework is further evaluated on SUSTech1K, CCPG, and Gait3D, where SMPL information is available or can be consistently estimated, achieving competitive performance under diverse real-world conditions.